
How to add an AI chat widget to your EmDash site
How I added a Talk To My Sites chat widget to rogerstringer.com, an EmDash site: one component in the Astro layout, lead capture, the questions dashboard, and Turnstile.
Read more
How I added a Talk To My Sites chat widget to rogerstringer.com, an EmDash site: one component in the Astro layout, lead capture, the questions dashboard, and Turnstile.
Read more
Two config files and four commands: the Hermes gateway and its WebUI on one Fly.io Machine, sharing a volume, with each container holding only the secrets it needs.
Read more
Most sites already have the answers, buried on the right paragraph of the right page. Talk To My Sites reads your sitemap, indexes every page, and gives you one script tag so visitors can ask and get answers from your own content, with sources.
Read more
Once four agents can edit the same repo, architecture stops being the hard problem and coordination starts. The five roles I run and what each one is forbidden to touch, why every handoff goes through a file, a worktree script so agents stop racing, and the failure modes worth naming.
Read more
Saying "done" costs a model nothing, which means it carries no information. A definition of done has four machine-checkable parts: source boundaries, acceptance criteria written as observations, verification commands the agent does not own, and a list of claims it must refuse to make without evidence.
Read more
A one-shot QA pass is a demo. Put it on a cron and it files the same issue every morning until the qa-bot label is a landfill. Here is the loop that makes the queue drain instead of grow: a spec in version control, fingerprinted issues, a reconciler with no opinions, and auto-close when the flow goes green.
Read more
A practical framework for giving production agents the right evidence, the right limits, and a measurable definition of done.
Read more
Most MCP tutorials assume a standalone process. Here's how to expose one as just another Astro API route using Streamable HTTP — same auth, same deploy pipeline, no second app to maintain.
Read more
Extending the Datastar chat widget with the browser's native SpeechRecognition API — no new backend, no transcription service, just one more way to fill the same signal.
Read more
Take the Astro + Datastar + Vercel AI SDK stack from this series and run it entirely on your own machine with Ollama — no API key, no per-token cost, no data leaving your laptop. It's a one-import change.
Read more
Stream typed, structured data from the Vercel AI SDK and watch a card build itself field by field in a Datastar UI — using streamText with output, the v6 replacement for streamObject. No React.
Read more
My dialer QAs itself: Claude Code logs in, dials a list of Twilio test numbers — a voicemail, an IVR, a simulated gatekeeper — records what happened vs. what should have, and files GitHub issues for another agent to fix.
Read more
Turn the streaming chat widget into an agent that calls tools — and stream the tool-call status live into the conversation. Astro, Datastar, and the AI SDK v6 ToolLoopAgent, no React.
Read more
A full multi-turn chat widget with live token streaming — where the server owns the conversation and the DOM is just the view. No React, no useChat, no client framework.
Read more
A small, genuinely useful build: a button that streams an AI summary of the current page into a panel, using Astro, Datastar, and the Vercel AI SDK — no React, no client framework.
Read more
The Vercel AI SDK's useChat only ships React, Vue, and Svelte bindings. Here's the small, reusable pattern for streaming AI tokens into a plain Datastar UI instead — no framework required
Read more
The Vercel AI SDK's useChat only ships React, Vue, and Svelte bindings. This six-part series builds real AI apps — chat, agents, structured output, local models — with no client framework at all, using Astro, Datastar, and one small streaming pattern
Read more
How I wired up a tiny TLDR tool that takes a URL, extracts the article, and streams a Markdown summary back to the browser in real time — using Astro SSR, Datastar for reactivity, and the Vercel AI SDK for the model call.
Read more
TLDR: Bun has been acquired by Anthropic. Anthropic is betting on Bun as the infrastructure powering Claude Code, Claude Agent SDK, and future AI coding products & tools.
Read more
n8n is a powerful workflow automation tool that helps connect different services and automate tasks, ideal for personal or small-scale use.
Read more
A browser-based AI-powered software creation platform called Replit appears to have gone rogue and deleted a live company database with thousands of entries. What may be even worse is that the Replit AI agent apparently tried to cover up its misdemeanors, and even ‘lied’ about its failures. The Replit CEO has responded, and there appears to have already been a lot of firefighting behind the scenes to rein in this AI tool.
Read more
OpenAI’s deal to buy Windsurf is off, and Google will instead hire Windsurf CEO Varun Mohan, cofounder Douglas Chen, and some of Windsurf’s R&D employees and bring them onto the Google DeepMind team, Google and Windsurf announced Friday.
Read more
Claude Opus 4 is the world’s best coding model, with sustained performance on complex, long-running tasks and agent workflows. Claude Sonnet 4 is a significant upgrade to Claude Sonnet 3.7, delivering superior coding and reasoning while responding more precisely to your instructions.
Read more
Vibe coding is great, using some sort of AI assistance to help code your app, but it isn't entirely as straight forward as some think. People tend to think they can just open up an IDE such as cursor or windsurf and ask it to build something but that's not how to achieve success with it.
Read moreBetween May and August, more AI Pins were returned than purchased, according to internal sales data obtained by The Verge. By June, only around 8,000 units hadn’t been returned, a source with direct knowledge of sales and return data told me. As of today, the number of units still in customer hands had fallen closer to 7,000, a source with direct knowledge said.
Read moreMicrosoft’s relationship with OpenAI has officially become more complicated.
Read moreHas AI frozen the gates of hell?
Read moreHumane is telling AI Pin owners today that they should “immediately” stop using the charging case that came with its AI gadget. There are issues with a third-party battery cell that “may pose a fire safety risk,” the company wrote in an email to customers.
Read moreHumane, the startup behind the poorly-reviewed AI Pin wearable computer, is already hunting for a potential buyer for its business. That’s according to a report from Bloomberg, which says the company — led by former longtime Apple employees Imran Chaudhri and Bethany Bongiorno — is “seeking a price of between $750 million and $1 billion.”
Read moreThe term "open" has no agreed-upon definition in the context of AI, and is applied to widely divergent offerings with little reference to a stable meaning.
Read more